Designed by Eero Simoncelli and Edward Adelson, July 1989.
EPIC is an experimental lossy image compression utility written in the C programming language. The compression algorithm is based on a critically-sampled non-orthogonal (imperfect-reconstruction) dyadic wavelet decomposition and a combined run-length/Huffman entropy coder.
EPIC is designed for extremely fast decoding on conventional (ie, non-floating point) hardware, at the expense of slower encoding and a slight degradation in compression quality (as compared to a good orthogonal wavelet decomposition).
EPIC compression quality is quite good (see images below), especially considering the simplicity of the algorithm. For more information, see the README file, our paper from the 1990 Picture Coding Symposium, our paper from the 1987 SPIE conference, my MS thesis, or a more general book chapter on subband transforms.
More recently (Spring, 1997), we released a much more powerful image compression algorithm called "EPWIC" (Embedded Predictive Wavelet Image Coder). A brief description, links to references, and C source code are available.
Full C source code for EPIC is available in a compressed UNIX tar file. The latest modifications to the program are described in the ChangeLog file.
We are making this code available to interested researchers who wish to experiment with a subband pyramid coder. These programs shall not be used, rewritten, or adapted as the basis of a commercial software or hardware product without first obtaining appropriate licenses from M.I.T.
Original (512x512) | Compressed 20:1 | Compressed 40:1 | Compressed 80:1 |
---|---|---|---|
262144 bytes | 13103 bytes | 6550 bytes | 3275 bytes |
![]() |
![]() |
![]() |
![]() |
Original (512x512) | Compressed 20:1 | Compressed 40:1 | Compressed 80:1 |
---|---|---|---|
262144 bytes | 13092 bytes | 6550 bytes | 3275 bytes |
![]() |
![]() |
![]() |
![]() |